From mboxrd@z Thu Jan 1 00:00:00 1970 From: Joe Thornber Subject: Re: dm-thinp bug Date: Wed, 27 Apr 2011 10:52:44 +0100 Message-ID: <1303897964.4679.56.camel@ubuntu> References: <20110426184742.GA13880@infradead.org> <1303895339.4679.51.camel@ubuntu> <20110427093259.GA24141@infradead.org> Reply-To: device-mapper development Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20110427093259.GA24141@infradead.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: dm-devel-bounces@redhat.com Errors-To: dm-devel-bounces@redhat.com To: Christoph Hellwig Cc: dm-devel@redhat.com, ejt@redhat.com List-Id: dm-devel.ids On Wed, 2011-04-27 at 05:32 -0400, Christoph Hellwig wrote: > > This patch already creashes when creating the table: Sigh, it's revealed a bug in io_hints, which isn't ever run if there's no iterate_devices callback. The patch below has been regression tested, but not tried above a virtio device. It's been pushed too. - Joe diff --git a/drivers/md/dm-thin-prov.c b/drivers/md/dm-thin-prov.c index 6ac802e..a168ee6 100644 --- a/drivers/md/dm-thin-prov.c +++ b/drivers/md/dm-thin-prov.c @@ -641,7 +641,15 @@ thinp_io_hints(struct dm_target *ti, struct queue_limits *limits) struct thinp_c *tc = ti->private; blk_limits_io_min(limits, 0); - blk_limits_io_opt(limits, data_dev_block_size(tc)); + blk_limits_io_opt(limits, tc->block_size << tc->block_shift); +} + +static int thinp_iterate_devices(struct dm_target *ti, + iterate_devices_callout_fn fn, + void *data) +{ + struct thinp_c *tc = ti->private; + return fn(ti, tc->data_dev, 0, tc->data_size << tc->block_shift, data); } /* Thinp pool control target interface. */ @@ -659,6 +667,7 @@ static struct target_type thinp_target = { .status = thinp_status, .merge = thinp_bvec_merge, .io_hints = thinp_io_hints, + .iterate_devices = thinp_iterate_devices, }; static int __init dm_thinp_init(void)